Targeted muscle reinnervation (TMR) is a powerful new tool in preventing and treating residual limb and phantom limb pain. In the adult population, TMR is rapidly becoming standard of care; however, there is a paucity of literature regarding indications and outcomes of TMR in the pediatric population. We present 2 cases of pediatric patients who sustained amputations and the relevant challenges associated with TMR in their cases. One is a 7-year-old patient who developed severe phantom and residual limb pain after a posttraumatic above-knee amputation. He failed pharmacologic measures and underwent TMR. He obtained complete relief of his symptoms and is continuing to do well 1.5 years postoperatively. The other is a 2-year-old boy with bilateral wrist and below-knee amputations as sequelae of sepsis. TMR was not performed because the patient never demonstrated evidence of phantom limb pain or symptomatic neuroma formation. We use these 2 cases to explore the challenges particular to pediatric patients when considering treatment with TMR, including capacity to report pain, risks of anesthesia, and cortical plasticity. These issues will be critical in determining how TMR will be applied to pediatric patients.
Although the most common etiology for major limb loss in children is congenital, acquired limb loss sufficient to require inpatient hospitalization related to trauma or cancer affects 1.3 to 6.3 children per 100 000.1–3 One of the main sequelae of major limb amputation in children is chronic pain.4,5 Symptomatic neuromas after amputation can lead to residual limb pain (RLP), which includes pain localized to the stump or irritation by external stimuli, such as temperature, pressure, or light touch. Phantom limb pain (PLP) is the sensation of discomfort in an amputated limb.6 For both neuromas and PLP, there is a wide range of interventions, from medications to spinal cord stimulators; however, efficacy has been inconsistent and often only incremental.7–10
Targeted muscle reinnervation (TMR) was initially developed to allow volitional control of a myoelectric prosthesis. In the procedure, the transected nerves are coapted to small intramuscular nerve branches (Fig 1). Clinical assessments of these patients unexpectedly revealed TMR also prevented or reduced RLP and PLP.11,12 Randomized controlled trials of adult patients with amputation support these observations and further demonstrated that TMR appears to be particularly effective at the time of amputation.6,13–15 Because of expanding evidence of positive clinical outcomes, several academic centers now consider TMR the standard of care for all elective and traumatic amputations.8,16 Although TMR is potentially useful in the future of bioprosthetics, most patients do not have access to these devices. A recent review indicated that ∼67 patients worldwide live with myoelectric upper limb prosthetics developed for patients with amputation who underwent TMR.17 In contrast, current literature reports at least 308 patients treated with TMR for neuropathic pain since 2014, thus illustrating the evolution for the primary use of TMR in patients with amputation.6,14–16 ,18–23 There has only been a single report of TMR in a child for a myoelectric prosthesis, and there have been no reports of TMR and effects on pain in children.24 Here, we present 2 cases exploring TMR application in the pediatric population and associated challenges.
Targeted muscle reinnervation (TMR) coapts the distal end of large mixed motor and sensory nerves in amputees to small intramuscular motor branches of remaining intact nerves. In a trans-humeral amputation, the median nerve is coapted to a branch of the musculocutaneous nerve. This allows median nerve signals to create an EMG signal in the muscle to facilitate prosthesis control. Increasing evidence also shows this procedure prevents and treats residual limb pain and phantom limb pain.
Targeted muscle reinnervation (TMR) coapts the distal end of large mixed motor and sensory nerves in amputees to small intramuscular motor branches of remaining intact nerves. In a trans-humeral amputation, the median nerve is coapted to a branch of the musculocutaneous nerve. This allows median nerve signals to create an EMG signal in the muscle to facilitate prosthesis control. Increasing evidence also shows this procedure prevents and treats residual limb pain and phantom limb pain.
Case 1
The patient is a 7-year-old boy who sustained a gunshot injury to the right flank. The pediatric surgical team performed an emergency laparotomy, revealing injuries to the left common, external, and internal iliac arteries. Despite vascular grafting, his foot and lower leg musculature progressed to necrosis and an above-knee amputation was performed 1 week after admission. Before the amputation, plastic surgery service was consulted to discuss TMR for pain prevention and future benefit with myoelectric prosthetic use. TMR was declined at the time because of absence of significant pain.
After discharge, the patient described his absent “toes wiggling at night.” This progressed to PLP disrupting his sleep. He was seen in the chronic pain clinic and was started on gabapentin and amitriptyline. At 3 months post amputation, despite multidisciplinary pain management, he continued to have significant PLP and RLP. He was referred back to plastic surgery, and they elected to proceed with TMR. The nerve transfers performed included the following: left common peroneal to motor branch to biceps femoris, tibial to motor branch to semimembranosus, posterior cutaneous nerve of the thigh to motor branch to semimembranosus, and saphenous to motor branch to vastus medialis (Fig 2). The patient’s postoperative course was uncomplicated, and he was discharged on postoperative day 1. One week later, the patient and his mother reported he was sleeping through the night and no longer requiring pain medication at night. Given his rapid improvement, his pain management team began to wean his gabapentin and amitriptyline. He had a brief flare of his night pain 2 months post operation, but this quickly resolved after a brief repeated course of gabapentin. By 4 months post operation, he was no longer taking medication, and his pain had entirely resolved. He was fitted with a prosthesis, and at the 18-month follow-up, he denied PLP and RLP.
TMR was performed through posterior and anterior approaches. A, The sciatic nerve was split into common peroneal and tibial branches and the red vessel loops are around the recipient nerves: motor branches to biceps femoris and semimembranosus. Additionally, the posterior cutaneous nerve to thigh was dissected and coapted to a branch to semimembranosus. B, Through an anterior approach the saphenous nerve was coapted to a branch to vastus medialis.
TMR was performed through posterior and anterior approaches. A, The sciatic nerve was split into common peroneal and tibial branches and the red vessel loops are around the recipient nerves: motor branches to biceps femoris and semimembranosus. Additionally, the posterior cutaneous nerve to thigh was dissected and coapted to a branch to semimembranosus. B, Through an anterior approach the saphenous nerve was coapted to a branch to vastus medialis.
Case 2
The patient is a 9-month-old boy admitted for septic shock, with cultures positive for Haemophilus influenzae. He developed disseminated intravascular coagulopathy and renal failure. He was treated with epinephrine, norepinephrine, angiotensin II, and dobutamine and, unfortunately, developed vasopressor-induced ischemia of all 4 extremities. He underwent bilateral wrist disarticulations and below-knee amputations at 11 months of age. Distal amputated nerves were tagged for possible future TMR, but this was not pursued during the initial admission because of the patient’s health status. He was fitted for prosthetics for each limb at 5 months post operation. At 7 months post operation, he had some discomfort with the lower extremity prosthetic but no pain with direct palpation or at rest. The patient and his family never reported pain at rest or from palpation throughout his postoperative course. Given this, the parents and treatment team felt TMR was not indicated, and additional surgery would delay prosthesis use.
Discussion
Amputation-related pain is understudied in the pediatric population, with most support for evidence-based treatments deriving from small case studies and a varied incidence of 12% to 83% being reported.4,25,26 Evidence-based treatment modalities include pharmacologic therapy (gabapentin27 and amitriptyline28 ), psychotherapy,29 acupuncture,30 and mirror therapy.31 The patient in case 1 continued to have significant PLP and RLP despite initial multimodal pain management, including medications and cognitive behavioral therapy. Based on our institution’s experience with adult patients with amputation, we felt our older patient would benefit from TMR. Although TMR is intricate and more expensive than other evidence-based treatment modalities, further investigation is required to weigh potential success versus risks, costs, labor intensity on patients, and degree of compliance between TMR and other interventions. Primary TMR (TMR performed at the time of amputation) may be preferred in the future for its preventive nature and diminished risk compared with secondary procedures. Still, there are unique factors to consider when applying TMR in the pediatric population, such as expression of pain, cortical plasticity, age-related anesthetic risk, and future prosthetic use.
PLP has both central and peripheral contributions. There is reorganization in the primary sensory-motor cortex and ongoing spontaneous afferent activity from the injured peripheral nerve.7 PLP appears to be rarer in younger children, which correlates with observations that PLP occurs at a lower rate in patients with congenital versus traumatic amputation. It is postulated that this may be due to enhanced adaptive neuroplasticity in younger children.4,25 Thus, there may be reduced risk of PLP in a 2-year-old compared with the 7-year-old patient with amputation. This pattern is further highlighted by case 1: despite delayed time to TMR (3 months), the patient had significant pain relief. Previous studies suggest adult patients with amputation may have better outcomes if TMR is performed at or near the time of amputation.6,8,15 This may indicate a longer period of peripheral nerve adaptability after injury in children compared with adults. It was interesting how the PLP so rapidly improved after TMR, similarly rapid analgesia has been found in an animal model of TMR, and this suggests that peripheral activity was the most significant contributor to PLP in our patient.32 The time from treatment to observed change in PLP has not been well studied in adults because most reported assessments were at the latest follow-up. Ultimately, this plasticity could offer children a more flexible treatment window, with time to explore other treatment modalities.
An additional consideration in pediatric surgery includes the potential risks associated with general anesthesia exposure. Despite conflicting literature, one study found that slightly less than 1 hour of general anesthesia in early infancy does not alter neurodevelopmental outcomes at age 5 years compared with awake-regional anesthesia.33–36 Although an appropriate concern when considering a secondary TMR surgery, as in case 1, the minimal risk associated with one-time anesthetic use may promote the indication for primary TMR in pediatric cases.
Another unique factor in pediatrics is the assessment of pain. Compared with pain assessment in adults, there are additional obstacles due to the variability in physical and psychological development, including comprehension and the ability to communicate pain.31,37–39 A standardized assessment tool for both patients and caregivers, specific for PLP and RLP, will be integral for determining the efficacy of TMR in pediatric patients. Concerning our cases, the older patient who underwent TMR could clearly communicate sensations of PLP and RLP. In contrast, pain assessment in the much younger patient in case 2 depended on physician examination and caregiver observations.
For growth and development, it is unlikely that TMR would have significant effects. However, there is concern that using small motor branches from muscles in the distal stump could result in loss of critical bulk. For this reason, it is advised in below-knee-amputations to only use muscle branches to only one head of the gastrocnemius to conserve distal bulk. The increased risk of terminal osseous overgrowth and eventual skin perforation in pediatric patients with amputation could make preserved bulk more critical.40
A final confounder in TMR application in pediatric patients is the original purpose: control of a myoelectric prosthesis.11,41 Many barriers to their use remain, including cost and access. For patients with below-knee amputation, myoprosthetics have less functional value; however, myoprosthetics have tremendous functional value for patients with upper extremity amputation.42 As the availability of prosthetics increases, it may become standard practice to perform TMR acutely for upper extremity myoprosthetics and potential pain benefits, whereas patients with lower extremity amputation may undergo TMR on an as needed basis for pain.
Conclusions
Evidence shows that TMR provides multiple benefits to patients with amputation: relief from PLP and RLP and capacity for myoelectric prosthetic use. The current literature almost exclusively relates to adult patients with amputation, so evidence for indications and outcomes of TMR in the pediatric population is currently insufficient. We compared a 7-year-old patient with above-knee amputation who achieved resolution of PLP and RLP with TMR with a 2-year-old patient with 4-limb amputation who never appeared to develop PLP or RLP. The differing patient factors led to a contrast in the decision-making process regarding TMR and demonstrated essential aspects one must consider before performing TMR on a pediatric patient.
FUNDING: No external funding.
Drs Hoben and Vial conceptualized and designed the study, drafted the initial manuscript, and reviewed and revised the manuscript; Ms Lieb conceptualized and designed the study, researched and drafted portions of the discussion, drafted the initial manuscript, and reviewed and revised the manuscript; Ms Pysick researched and drafted portions of the discussion; Drs Rusy and Hettinger conceptualized and designed the study and critically reviewed and revised the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
Comments